This study examines the effects of temperature and precipitation on the mean and variance of seasonal rice yield in Andhra Pradesh, India, over a period of 33 years (1969-2002). For this purpose, two distinct approaches are employed: (i) panel data analysis using Just and Pope stochastic production function and (ii) quantile regression approach. The first approach suggests that, in general, an increase in temperature as well as inter-annual variance of temperature and rainfall adversely affect the mean crop yield, while the effect of increase in precipitation highly depends on the cropping season. Furthermore, an increase in average temperature, rainfall and their respective interannual variance are likely to increase inter-annual variability in crop yield. Second, the quantile regression reveals that rice yield's sensitivity to climate change differs signi cantly across the quantiles of yield distribution. In particular, the adverse effect of climate change is found to be more profound for the crop yields in lower quantiles. In addition, evidences in support of heterogeneity in the impact of climate change across the agro-climatic zones are also found. Overall, these findings call for a more location speci c adaptation policies to address heterogeneity and an integrated policy framework covering the downside risk to build resilience in the food security system.